https://orcid.org/0000-0001-5704-343X
ABSTRACT
Although previous studies have shown no increased mortality risk after the primary series of COVID-19 mRNA vaccines, reports on booster doses are lacking. This study aimed to evaluate mortality risk after the mRNA vaccine boosters in addition to the primary series. This nested case-control study included two age-specific cohorts (18–64 and ≥65 years as of February 1, 2021) in two municipalities. All deaths were identified and matched five controls for each case at each date of death (index date) with risk set sampling according to municipality, age, and sex. The adjusted odds ratios (aORs) and 95% confidence intervals (CIs) for mRNA vaccines (first to fifth doses) were estimated by comparing with no vaccination within 21 and 42 days before the index date using a conditional logistic regression model. The 18–64-years cohort comprised 431 cases (mean age, 57.0 years; men, 58.2%) and 2,155 controls (mean age, 56.0; men, 58.2%), whereas the ≥65-years cohort comprised 12,166 cases (84.0; 50.2%) and 60,830 controls (84.0, 50.2%). The aORs (95% CI) in 0–21 days after the third and fourth doses in the 18–64-years cohort were 0.62 (0.24, 1.62) and 0.38 (0.08, 1.84), respectively. The aORs (95% CI) after the third to fifth doses in the ≥65 years cohort were 0.36 (0.31, 0.43), 0.30 (0.25, 0.37), and 0.26 (0.20, 0.33), respectively. In conclusion, booster doses of mRNA vaccines do not increase mortality risk. These findings could help subsequent vaccine campaigns and alleviate vaccine hesitancy.
Introduction
A vaccination program against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was initiated in Japan on February 17, 2020, mainly targeting healthcare practitioners. From April 12, 2021, the program was expanded to high-risk populations such as those with older age (≥65 years), with underlying diseases, or working in long-term care facilities. The approved coronavirus disease (COVID-19) vaccines include BNT162b2, mRNA-1273, bivalent vaccines (ancestral strain and BA.1 or BA.4/5), ChAdOx1 nCoV-19, Ad26.COV2.S, and NVX-CoV2373. In total, 382,594,909 doses have been administered by March 13, 2023,Citation1 including 86,210,270; 58151,216; and 30,215,678 of third, fourth, and fifth doses of vaccination, respectively.
A continued vaccination program with mRNA vaccines is planned in Japan from September 2023 onward, based on the current efficacy/effectiveness and safety profiles of these products. Previous studies have demonstrated the high efficacy and effectiveness of mRNA vaccines for COVID-19.Citation2–4 Rapid safety assessments between vaccinations and adverse events of concern in real-world settings have contributed to the establishment of the safety profiles of these vaccines.Citation5,Citation6 The currently recognized adverse reactions to mRNA vaccination include both common (headache, fatigue, fever, muscular pain, and chills) and rare (anaphylaxis, myocarditis, and pericarditis) adverse reactions.Citation7,Citation8
Deaths following mRNA vaccination have been documented, irrespective of any causal relationship between the vaccination and mortality.Citation9–11 In Japan, 2,053 deaths in individuals aged ≥12 years after mRNA vaccination have been reported by March 12, 2023 (numbers of reported cases per 1,000,000 doses: 6.2 and 2.7 for BNT162b2 and mRNA-1273, respectively).Citation10,Citation11 However, a case-by-case qualitative causal assessment of 2,041 reported cases was not possible because of a lack of information. Evaluations of the existing cases revealed that 11 did not establish a causal relationship between vaccination and death, whereas, in one instance, such a relationship could not be ruled out. Consequently, a quantitative assessment of mortality risks following mRNA vaccination is imperative for appraising the benefits and drawbacks of future vaccination programs.
Previous studies have shown no significant increase in mortality risk after the first and second doses of COVID-19 vaccine.Citation12–19 The study, which assessed the mortality risk after the first and second doses of monovalent mRNA vaccine using a cohort and self-controlled case series design in one municipality in Japan, reported incident rate ratios of all-cause mortality after the first and second doses of 0.32 (0.25–0.41) and 0.32 (0.25–0.42), respectively.Citation18 However, these studies only assessed mortality risks after the primary series; thus, studies on booster doses are required. Therefore, the present study aimed to assess the mortality risks after the third, fourth, and fifth doses of mRNA vaccines in addition to the primary series.
Materials and methods
Design, setting, and data sources
We conducted a nested case-control study, which is one type of case-control study carried out within a cohort.Citation20 This nested case-control study analyzed data obtained from the Vaccine Effectiveness, Networking, and Universal Safety (VENUS) Study.Citation21 We used data from two municipalities (Kanto and Chubu region) in Japan between January 1, 2019 and December 31, 2022. The data included individuals covered by the National Health Insurance or Latter-Stage Older Persons Health Care System. Therefore, most participants in this study were older adults (age ≥65 years), as well as self-employed individuals and their dependents. The database included data from the Vaccination Record System, Health Center Real-time Information-Sharing System on COVID-19 (HER-SYS), Basic Resident Register, and healthcare claims. SARS-CoV2 test-positive cases were registered in the HER-SYS until September 25, 2022, with only high-risk patients with COVID-19 reported from September 26, 2022. This study was approved by the Kyushu University Institutional Review Board for Clinical Research (no. 22114–00). The requirement for individual informed consent was waived based on the Japanese ethical guidelines, as this secondary analysis used routinely collected anonymized data by the municipalities.
Cohort, case, and control selection
We defined February 17, 2021, the day on which mRNA vaccination was started in Japan, as the cohort entry date and the two age-based cohorts as 18–64 and ≥65 years. Therefore, individuals had the chance to obtain vaccination doses from the above date. The study included individuals who met the following criteria: 1) age 18–64 or ≥65 years as of February 1, 2021, and 2) had available healthcare claims before the cohort entry date to identify individuals covered by the National Health Insurance or Latter-Stage Older Persons Health Care System. Individuals who met the following criteria were excluded: 1) confirmed SARS-CoV2 test positive before the cohort entry date and 2) those who died or moved from the municipality before the cohort entry date. Participants in this study were followed up until the following dates: 1) event date (death); 2) before testing positive for SARS-CoV-2 to exclude the effect of the infection; 3) moved to another municipality because we could not obtain their information on the outcomes and the exposures; 4) before vaccination with another COVID-19 vaccine types that were not targeted in the study (ChAdOx1 nCoV-19 or NVX-CoV2373); and 5) end of the study period (18–64-years cohort: September 25, 2022, because the population did not mandatorily report cases of COVID-19; ≥65-years cohort: December 31, 2022), whichever came first.
We identified cases and controls from the cohort by risk set sampling with replacement.Citation20 Cases were identified from death records between the follow-up periods, and the calendar date of death was defined as the index date. The controls were matched at a 1:5 ratio based on municipality, age (5-year groups), and sex. The index date for each control was defined as the same calendar date as the index date for the matched case. The risk set sampling with replacement was a matching process conducted by randomly selecting five controls from the risk set (among all individuals considered to be at risk) at the time points when a case was identified. If an individual transitioned from being a control to a case, they served both as a control and a case at each relevant time point.
Exposure and covariates
For the primary analysis aimed at assessing the risk associated with each dose, we designated each dose – be it first, second, third, fourth, or fifth – as falling within a period of 21 to 42 days prior to the index date. The individuals who did not meet the exposure criteria were categorized as unexposed regardless of their vaccination status. When several (first and second) doses were identified within the period before the index date, the exposure was defined as the latest dose. The 18–64-years and ≥65-year cohorts were assessed until the fourth and fifth doses, respectively, due to the vaccination schedule in Japan and our study period based on the infection reporting policy. The vaccination program was scheduled, prioritizing older and high-risk populations. The fourth dose started on March 25, 2022 for the population vaccinated with the third dose more than five months before. The start date of the fifth dose vaccination was not publicly announced by the Japanese government; however, the fifth dose vaccination has been reported since October 5, 2022. On October 22, 2022, the eligibility was updated to the population vaccinated with last booster doses more than three months before. The reporting patients with COVID-19 were limited to those aged ≥65 years, who had risk factors for severe COVID-19, who required anti-COVID-19 medication or supplementary oxygen, or who were pregnant from September 26, 2022. For secondary analysis to assess the risk of any doses, we defined any dose administered within 21 and 42 days before the index date. This secondary analysis could estimate the morality risk in populations with a low event rate, particularly in the 18–64-year cohort.
We assessed the covariates of age and sex at the cohort entry date, comorbidities (myocardial infarction, congestive heart failure, peripheral vascular disease, cerebrovascular disease, dementia, chronic pulmonary disease, rheumatic disease, peptic ulcer disease, mild liver disease, diabetes without and without chronic complications, hemiplegia or paraplegia, renal disease, any malignancy, moderate or severe liver disease, metastatic solid tumor, and AIDS/HIV) 1 year before the cohort entry date. Additionally, we considered the Charlson comorbidity index.Citation22
Statistical analysis
We summarized the demographic data (age and sex) in the case and control groups. To assess mortality risk after COVID-19 vaccination, we performed conditional logistic regression with 17 diseases as covariates to estimate the odds ratios (ORs) with 95% confidence intervals (95% CIs). All analyses were performed using R version 4.1.2 (R Foundation for Statistical Computing, Vienna, Austria).
Results
A total of 83,730 and 186,489 individuals aged 18–64 and ≥65 years, respectively, with previous claims before February 2021, were identified in the two municipalities from the VENUS Study (). Among them, the 18–64- and ≥65-years cohorts included 76,730 and 171,967 participants, respectively. In the 18–64-years cohort, 431 deaths were identified in 114,263 patient-years, whereas in the ≥65-years cohort 12,170 deaths were identified in 303,309 patient-years. Using risk-set sampling, 431 cases and 2,155 controls (aged 18–64 years) and 12,166 cases and 60,830 controls (aged ≥65 years) were matched.
Figure 1. Study flow chart.
In the 18–64-years cohort, the median ages (interquartile range [IQR]) were 57.0 (49.0, 61.0) years for cases and 56.0 (49.0, 61.0) years for controls (). For the ≥65-years cohort, the median ages (IQR) in the cases and controls were 84.0 (78.0, 90.0) and 84.0 (78.0, 90.0) years, with 6,110 (50.2%) and 30,550 (50.2%) men, respectively. The proportions of each comorbidity differed between the cases and controls in the 18–64-years cohort. Dementia, hemiplegia or paraplegia, renal disease, any malignancy, moderate or severe liver disease, and metastatic solid tumor were especially frequent in cases than in controls in the ≥65-years cohort.
Table 1. Characteristics of the matched cases and controls.
The associations between mRNA vaccines and mortality in both cohorts are shown in . The adjusted ORs for the first to fourth doses among the 18–64-years cohort were similar to those in the ≥65-years cohort. Additionally, no increase in mortality risk was observed for each dose in 0–42 days, any doses in 0–21 days, and any doses in 0–42 days before the index date.
Table 2. Mortality risk after mRNA COVID-19 vaccination in the 18–64-years-old population.
Table 3. Mortality risk after mRNA COVID-19 vaccination in the ≥65-years-old population.
Discussion
This nested case-control study evaluated mortality risk after mRNA vaccination in two municipalities in Japan based on the VENUS Study. Our results showed no increase in mortality risk after the first (OR 0.67), second (OR 0.38), third (OR 0.62), and fourth (OR, 0.38) doses in adults aged 18–64 years, although we could not assess the fifth dose due to the vaccination schedule and study design. In the older cohort aged ≥65 years, mortality risk also did not increase after the first (OR 0.29), second (OR 0.31), third (OR 0.36), fourth (OR 0.30), and fifth (OR 0.26) doses.
The results in this study after vaccination were comparable to those reported in previous studies. A previous study reported incidence rate ratios of all-cause mortality of 0.27 and 0.26 after the first and second doses of mRNA vaccines, respectively.Citation18 Among nursing home residents, the risk ratios for 7-day mortality following the administration of the first and second doses of mRNA vaccines were 0.34 and 0.49, respectively.Citation14 The adjusted hazard ratios for 30-day non-COVID-19 mortality were 0.27 and 0.30 after the first and second doses of BNT162b2, and 0.19 and 0.25 after the first and second doses of mRNA-1273, respectively, in a population aged ≥12 years.Citation13 In another study, the crude hazard ratio of death among an older population aged ≥65 years was 0.28 at 21 days after the first dose of mRNA vaccines.Citation17 Thus, our results showed the same trends as those of previous studies. Our results also provided additional safety information for booster doses (up to the fifth dose) with the mRNA vaccine.
Low ORs (<1) between mortality and death after vaccination in our cohort study could be explained by healthy vaccinee bias,Citation23 although we adjusted for comorbidities as confounding factors. Healthier individuals are more likely to receive COVID-19 vaccines than individuals who are not in good condition. A cohort study that considered cognitive function and activities of daily living in addition to comorbidities using inverse probability weighting methods to assess 7-day mortality after vaccination reported estimated risk ratios < 1.Citation14 Although a self-controlled case series design could control for time-invariant confounders, previous studies using this design also reported low ORs.Citation18 We adopted a nested-case control design, considering the confounders and different exposure categories from those studies, to address various vaccination statuses during the study period; however, low ORs of primary and booster doses were estimated. Other factors not considered in previous studies that could reflect and relate to health conditions or short-term mortality should be identified in future studies to adjust for confounders.
Vaccination campaigns against SARS-CoV-2 infection are ongoing in Japan in 2023. Between May 8 and August 31, 2023, older adults aged ≥65 years, people with specific underlying diseases, and healthcare workers have been targeted for vaccination with bivalent mRNA vaccines. After September 20, 2023, individuals aged ≥6 months were eligible for vaccination with a monovalent XBB-containing vaccine. The proportion of anti-nucleocapsid antibodies induced by previous SARS-CoV-2 infection in blood donors has increased from previous reports; however, it was 42.8% overall. Thus, the population with infection-induced immunity is low compared with that of other countries.Citation24 COVID-19 vaccines are particularly important for high-risk populations in Japan. The findings of this study will contribute to an updated safety profile, alleviate concerns regarding the safety of COVID-19 vaccines, and encourage appropriate vaccination.
The current study had several limitations. First, residual confounding factors, such as health-seeking behaviors and detailed health status, were likely present. Second, this study represented only a part of the Japanese population because the study cohort only included two municipalities. In addition, only individuals covered by the National Health Insurance or Latter-Stage Older Persons Health Care System were included. In particular, the National Health Insurance mainly insures self-employed individuals and their dependents. Therefore, it may be difficult to generalize these results to young adults. Third, we could not assess the causes of death because we could not obtain this information, although death after SARS-CoV-2 infection was excluded from the study. Fourth, we could not assess the XBB.1.5 monovalent vaccines that will be used from September 2023 in Japan. Fifth, we could not guarantee a 1-year membership before the cohort entry date on membership and withdrawal although we confirmed previous claims. Therefore, the covariates of comorbidities might be misidentified. Sixth, conducting an analysis for each dosage within the 18–64-year age cohort would likely be underpowered in detecting associations, resulting in a wide confidence interval attributable to the limited sample size.
Conclusion
The first to the fourth doses of mRNA COVID-19 vaccination do not increase mortality risk. These findings provide additional safety information regarding mRNA COVID-19 vaccines, including bivalent vaccines, which could help subsequent vaccine campaigns and alleviate vaccine hesitancy.
Author contributions
WM developed the original protocol, and all authors reviewed and edited the protocol. MM, FM, and HF collected data. CI performed the data analysis. All authors interpreted the results of the analyses. WM drafted the original manuscript. All authors have reviewed and edited the manuscript. All the authors have read the manuscript and approved its submission for publication.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
Data cannot be made available for privacy or ethical reasons.
Additional information
Funding
References
- Cabinet Public Affairs Office, Cabinet Secretariat. COVID-19 vaccines; [accessed 2023 July 20]. https://japan.kantei.go.jp/ongoingtopics/vaccine.html.
- Graña C, Ghosn L, Evrenoglou T, Jarde A, Minozzi S, Bergman H, Buckley BS, Probyn K, Villanueva G, Henschke N, et al. Efficacy and safety of COVID-19 vaccines. Cochrane Database Syst Rev. 2022;12:CD015477. doi:10.1002/14651858.CD015477.
- Dagan N, Barda N, Kepten E, Miron O, Perchik S, Katz MA, Hernán MA, Lipsitch M, Reis B, Balicer RD, et al. BNT162b2 mRNA COVID-19 vaccine in a nationwide Mass vaccination setting. N Engl J Med. 2021;384(15):1412–6. doi:10.1056/NEJMoa2101765.
- Polack FP, Thomas SJ, Kitchin N, Absalon J, Gurtman A, Lockhart S, Perez JL, Pérez Marc G, Moreira ED, Zerbini C, et al. Safety and efficacy of the BNT162b2 mRNA COVID-19 vaccine. N Engl J Med. 2020;383(27):2603–15. doi:10.1056/NEJMoa2034577.
- Barda N, Dagan N, Ben-Shlomo Y, Kepten E, Waxman J, Ohana R, Hernán MA, Lipsitch M, Kohane I, Netzer D, et al. Safety of the BNT162b2 mRNA COVID-19 Vaccine in a nationwide setting. N Engl J Med. 2021;385(12):1078–90. doi:10.1056/NEJMoa2110475.
- Klein NP, Lewis N, Goddard K, Fireman B, Zerbo O, Hanson KE, Donahue JG, Kharbanda EO, Naleway A, Nelson JC, et al. Surveillance for adverse events after COVID-19 mRNA vaccination. JAMA. 2021;326(14):1390. doi:10.1001/jama.2021.15072.
- Husby A, Hansen JV, Fosbøl E, Thiesson EM, Madsen M, Thomsen RW, Sørensen HT, Andersen M, Wohlfahrt J, Gislason G, et al. SARS-CoV-2 vaccination and myocarditis or myopericarditis: population based cohort study. BMJ. 2021;375:e068665. doi:10.1136/bmj-2021-068665.
- CDC. Selected adverse events reported after COVID-19 vaccination; 2023 [accessed 2023 July 4]. https://www.cdc.gov/coronavirus/2019-ncov/vaccines/safety/adverse-events.html.
- Rosenblum HG, Gee J, Liu R, Marquez PL, Zhang B, Strid P, Abara WE, McNeil MM, Myers TR, Hause AM, et al. Safety of mRNA vaccines administered during the initial 6 months of the US COVID-19 vaccination programme: an observational study of reports to the vaccine adverse event reporting system and v-safe. Lancet Infect Dis. 2022;22(6):802–12. doi:10.1016/S1473-3099(22)00054-8.
- Ministry of Health, Labour and Welfare. Summary of cases reported as deaths following COVID-19 vaccination (spikevax) [In Japanese]; 2023 [accessed 2023 July 4]. https://www.mhlw.go.jp/content/10601000/001092258.pdf.
- Ministry of Health, Labour and Welfare. Summary of cases reported as deaths following COVID-19 vaccination (comirnaty) [In Japanese]; 2023 [accessed 2023 July 4]. https://www.mhlw.go.jp/content/10601000/001092112.pdf.
- Xu S, Huang R, Sy LS, Glenn SC, Ryan DS, Morrissette K, Shay DK, Vazquez-Benitez G, Glanz JM, Klein NP, et al. COVID-19 vaccination and non–COVID-19 mortality risk — Seven Integrated Health Care Organizations, United States, December 14, 2020–July 31, 2021. MMWR Morb Mortal Wkly Rep. 2021;70:1520–4. doi:10.15585/mmwr.mm7043e2.
- Xu S, Huang R, Sy LS, Hong V, Glenn SC, Ryan DS, Morrissette K, Vazquez-Benitez G, Glanz JM, Klein NP, et al. A safety study evaluating non-COVID-19 mortality risk following COVID-19 vaccination. Vaccine. 2023;41(3):844–54. doi:10.1016/j.vaccine.2022.12.036.
- Bardenheier BH, Gravenstein S, Blackman C, Gutman R, Sarkar IN, Feifer RA, White EM, McConeghy K, Nanda A, Mor V. Adverse events following mRNA SARS-CoV-2 vaccination among U.S. nursing home residents. Vaccine. 2021;39(29):3844–51. doi:10.1016/j.vaccine.2021.05.088.
- Stivanello E, Beghelli C, Cardoni F, Giansante C, Marzaroli P, Musti MA, Perlangeli V, Todeschini R, Pandolfi P. Short-term mortality following COVID-19 vaccination in Bologna, Italy: a one-year study. Vaccine. 2022;40(39):5709–15. doi:10.1016/j.vaccine.2022.08.039.
- Wong CKH, Lau KTK, Xiong X, Au ICH, Lai FTT, Wan EYF, Chui CSL, Li X, Chan EWY, Gao L, et al. Adverse events of special interest and mortality following vaccination with mRNA (BNT162b2) and inactivated (CoronaVac) SARS-CoV-2 vaccines in Hong Kong: a retrospective study. PLOS Med. 2022;19(6):e1004018. doi:10.1371/journal.pmed.1004018.
- Lopez-Doriga Ruiz P, Gunnes N, Michael Gran J, Karlstad Ø, Selmer R, Dahl J, Bøås H, Aubrey White R, Christine Hofman A, Hessevik Paulsen T, et al. Short-term safety of COVID-19 mRNA vaccines with respect to all-cause mortality in the older population in Norway. Vaccine. 2023;41(2):323–32. doi:10.1016/j.vaccine.2022.10.085.
- Takeuchi Y, Iwagami M, Ono S, Michihata N, Uemura K, Yasunaga H. A post-marketing safety assessment of COVID-19 mRNA vaccination for serious adverse outcomes using administrative claims data linked with vaccination registry in a city of Japan. Vaccine. 2022;40(52):7622–30. doi:10.1016/j.vaccine.2022.10.088.
- Nafilyan V, Bermingham CR, Ward IL, Morgan J, Zaccardi F, Khunti K, Stanborough J, Banerjee A, Doidge JC. Risk of death following COVID-19 vaccination or positive SARS-CoV-2 test in young people in England. Nat Commun. 2023;14(1):1541. doi:10.1038/s41467-023-36494-0.
- Vandenbroucke JP, Pearce N. Case-control studies: basic concepts. Int J Epidemiol. 2012;41(5):1480–9. doi:10.1093/ije/dys147.
- Fukuda H, Maeda M, Murata F. Development of a COVID-19 vaccine effectiveness and safety assessment system in Japan: the VENUS study. Vaccine. 2023;41(23):3556–63. doi:10.1016/j.vaccine.2023.03.059.
- Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi J-C, Saunders LD, Beck CA, Feasby TE, Ghali WA, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130–9. doi:10.1097/01.mlr.0000182534.19832.83.
- Remschmidt C, Wichmann O, Harder T. Frequency and impact of confounding by indication and healthy vaccinee bias in observational studies assessing influenza vaccine effectiveness: a systematic review. BMC Infect Dis. 2015;15(1):429. doi:10.1186/s12879-015-1154-y.
- Ministry of Health, Labour and Welfare. Preliminary reports of third survey of SARS-CoV-2 seroprevalence in blood donors in Japan [In Japanese]; 2023 [accessed 2023 July 20]. https://www.mhlw.go.jp/content/10900000/001108939.pdf.